Howe Sound Glass Sponge Reef Identification
By Lora McAuley, M.Sc, RPBio1
On behalf of the
Marine Life Sanctuaries Society
For
Fisheries and Oceans Canada
June 3rd, 2017
1
Marine Life Sanctuaries Society
Table of Contents
Introduction ................................................................................................ 1
Study Area .................................................................................................. 4
Data Collection and Mapping ...................................................................... 4
Bathymetry Mapping .......................................................................................................................................4
Imagery Data Collection ...................................................................................................................................5
Other Sources of Information ..........................................................................................................................5
Data Analysis ....................................................................................................................................................5
Results ........................................................................................................ 7
Nature and Landform .......................................................................................................................................7
Species Observations .......................................................................................................................................9
Assessment Summary ................................................................................................................................... 11
Area 1: Defence Islands / Ellesmere Creek ................................................................................................... 14
Area 2: Anvil Island ....................................................................................................................................... 18
Area 4: Halkett Point, Gambier Island .......................................................................................................... 26
Area 5: South Bowyer Island ......................................................................................................................... 28
Area 6: Dorman Point, Bowen Island ............................................................................................................ 31
Discussion ................................................................................................. 33
Acknowledgements .................................................................................. 36
References ................................................................................................ 37
Marine Life Sanctuaries Society 1
Introduction
Ancient and unique sponges of the Class Hexactinellida are found on the submerged reefs, pinnacles and
rock walls in Howe Sound. These are glass sponges, named for the internal silica framework, or skeleton,
that elevates the sponge above the seabed so it can feed, grow, and reproduce. Dense growths, or
aggregations, of glass sponge can transform a relatively homogenous reef into a complex and unique marine
ecosystem. They provide structural habitat for new generations of glass sponge and a refuge and foraging
ground for other species. They filter vast quantities of seawater for the bacteria that nourish them and expel
nutrients that, when upwelled to the photic zone, contribute towards photosynthesis (Kahn et al., 2015).
Unfortunately, the delicate nature and slow growth rate of glass sponges make them highly vulnerable to
damage from physical, biological and chemical stressors (Conway, 1999; Kahn et al., 2016). The damage or
death of a sponge aggregation impacts the entire ecosystem that depends on it, including species important
to aboriginal, recreational, and commercial fisheries in British Columbia.
In recognition of the importance of glass sponge ecosystems and their vulnerability, efforts have been made
to map and monitor aggregations along the coast of British Columbia. This endeavor can be challenging,
given the deep and often remote habitats that support glass sponges. Methods have involved the use of
multi-beam swath bathymetry, Remotely Operated Vehicle (ROV), and side-scan sonar (Conway et al., 2005;
Conway et al., 2007). In the Strait of Georgia and Howe Sound, mapping and monitoring has resulted in
bottom-contact fisheries closures of nine glass sponge reef complexes since 2015 (Fisheries and Oceans
Canada, in press). Recently, the Marine Life Sanctuaries Society (MLSS) and the Vancouver Aquarium have
revealed additional areas of significant glass sponge aggregations in Howe Sound that remain unprotected
from bottom contact fisheries.
Since 2012, the MLSS has mapped and monitored glass sponge aggregations in Howe Sound using drop
camera surveys and SCUBA (Clayton & Dennison, in press). In addition, the Vancouver Aquarium actively
collects data on physical, chemical and biological attributes from several of these aggregations (Vancouver
Aquarium, 2017). This has resulted in a wealth of information on Howe Sound glass sponge locations,
biological communities and general health. To date, there are 19 known glass sponge aggregations in Howe
Sound1, including those already protected from DFO closures to bottom contact fisheries. They are located
throughout Howe Sound, from Queen Charlotte Channel in the south to Ellesmere Creek Pinnacle in the
north (Figure 4).
Discoveries of glass sponge aggregations in Howe Sound align with current initiatives of the federal
government to identify and protect important ecological and biological marine benthic habitats. Fisheries
and Oceans Canada (DFO) is exploring long-term conservation measures along the coast of British Columbia
as part the Government of Canada’s international commitment to protect 5% of Canada’s marine coastline
by 2017 and 20% by 2020. As part of the Government of Canada’s commitments to conserve marine areas
and conduct fisheries in a sustainable manner, DFO commissioned the MLSS to summarize this information
for future use in the identification of management measures for Howe Sound glass sponge aggregations.
This report provides a synopsis of the status of knowledge and locations on glass sponge reefs in Howe
Sound that are not currently under the protection of DFO glass sponge fishery closures.
1 Six individual aggregations are encompassed within the two areas already protected with no-bottom contact fishery closures in Howe Sound.
Marine Life Sanctuaries Society 2
Glass sponge biology and ecology
Members of the sponge phylum Porifera share a general feeding and growth strategy that involves filtering
food from the water column through their body tissues, and expelling the unconsumed particles and wastes
through their exhalent chamber, or osculum (Pechenik, 1996). Like many sponges, the glass sponge body is
made up of a matrix of living cells supported by a spicule skeleton. Unlike other sponges, the spicules are
siliceous and six-rayed. Reef-forming glass sponges are further distinguished from other glass sponges in
that the spicules are fused, rather than loose (Leys, Mackie, & Reiswig, 2007). Upon death of the sponge, the
fused skeleton provides a scaffold on which new generations of sponge larvae can settle and grow (Conway
et al., 2001; Krautter et al., 2001). Over time, large biogenic reefs made entirely of glass sponge can develop.
The living cells of glass sponges are also unique from other sponges in that they lack a cell wall; instead they
are fused into a single mass of multinucleated tissue known as the syncytium that allows for the passage of
nutrients and electrical signals throughout the sponge (Leys et al., 2007; Wyeth et al., 1996). This feature
allows the sponge to quickly arrest feeding when exposed to adverse conditions such as sedimentation or
mechanical impacts (Tompkins-Macdonald & Leys, 2008).
In Howe Sound, glass sponge aggregations occur as bioherms (sponge growth on dead sponge) and as
gardens (sponge growth on rock). The bioherms are made up of two reef-building species of Hexactinellids
in the Family Aphrocallistidae. Aphrocallistes vastus, or the Cloud sponge, has a convoluted body with a high
surface area from the bends, folds, mittens, and spokes that branch from the main exhalent column, or
“chimney” (Figure 1A). Many oscula, or chimneys, may extend from a common base giving rise to large
mounds that reach up to 1.2m into the water column (Conway et al., 2005). Rarer is Heterochone calyx, or
the Goblet sponge, which typically has a wide oscular opening that tapers to a narrow base (Figure 1B). H.
calyx may be supported by basal projections of sponge tissue that anchor it to the substrate (Reiswig, 2002).
Figure 1: The two species of reef-forming glass sponge found in Howe Sound, British Columbia. A) The dominant reef builder Aphrocallistes vastus, or Cloud Sponge; and B) the more rare Heterochone calyx, or Goblet Sponge. In photo B, the diver is pointing at a squat lobster (Munida quadrispina), a common denizen of Howe Sound glass sponge reefs. Photos courtesy of Adam Taylor.
Sponge gardens may be formed by Rhabdocalyptus dawsoni, the Boot sponge, as well as A. vastus and H.
calyx, as well as (Figure 2). Boot sponges are tubular with an outer body of rough spicules that lead to a
large inner chamber with a wide osculum. The spicule arrangement in boot sponges is loose, and therefore
this species lacks the structural integrity to form bioherms.
Marine Life Sanctuaries Society 3
Glass sponge aggregations in Howe Sound occur in deep habitats exposed to currents, cool water, high silica
concentrations, and an abundant food supply (Whitney et al., 2005). Runoff from the Squamish River
contributes large amounts of terrestrially derived silica, which is eventually concentrated at depth where
sponges can utilize it for spicule development (Leys et al., 2004). High sediment runoff, temperature and
salinity stratification, and dense plankton blooms interact to inhibit mixing of warm surface layers to depths
where bioherms thrive (Stockner et al., 1977), likely helping to maintain temperatures within tolerance
ranges. This results in a highly productive upper layer and high settlement of sediments, nutrients, and
detritus below. Sponge aggregations can benefit from the settlement of food organisms and nutrients,
though sedimentation can be problematic if it interferes with their feeding (Leys, 2013).
Exposure to currents may play an important role in the distribution of glass sponges (Conway et al., 2005;
Whitney et al., 2005). They are often found atop pinnacles, submarine knolls and ridges where deep water
currents are accelerated. Currents help minimize the settlement of sediments on the sponge, yet still deliver
an abundant supply of food and nutrients (Leys, 2013). Currents also influence the energetics associated
with feeding by facilitating filtration (Leys et al., 2011). In the absence of external currents, food is drawn
into the sponge body and the wastes expelled by the actions of flagellated collar cells, or choanocytes. The
beating of the flagella requires oxygen, which is often depleted at depths where glass sponges occur. This
problem is alleviated in high current situations where the water moves passively through the sponge,
reducing the energy and oxygen required by the flagella.
Glass sponges in the Strait of Georgia occupy a vertical distribution that corresponds to a narrow
temperature range between 9.4-10.5 °C (Leys et al., 2004). Outside this range, the ability for the syncytial
tissue to transmit an electrical signal is impeded. Vertical distribution in glass sponges may also be
influenced by the availability of dissolved silica, a nutrient essential in the formation of the silicon dioxide
skeleton. Dissolved silica is often less concentrated in the shallows than at depth, likely due to the uptake by
diatoms in shallow waters (Chu et al., 2011). Deeper areas with higher dissolved silica concentrations often
coincide with the presence of sponge aggregations along the coast of British Columbia, indicating this is
important for sponge growth (Austin, 1999; Chu et al., 2011; Leys et al., 2004; Yahel et al., 2007).
Figure 2: Example of a sponge garden, or sponge growth on rock, with the Boot sponge, Rhabdocalyptus dawsoni, to the left and the more colourful A. vastus to the right. Photo courtesy of Adam Taylor.
Marine Life Sanctuaries Society 4
The delicate nature of the sponge skeleton makes them susceptible to breakage upon physical impact.
Although they have the capacity to recover from damage once the stressor is removed, growth is slow
(Austin et al., 2007; Marliave, 2015). Reported growth rates of glass sponges typical of those found in Howe
Sound range from 1-6cm/year for R. dawsoni (Leys & Lauzon, 1998) and up to 10 cm/year for A. vastus
(Austin et al., 2007). However, damage to the underlying substrate of sponge skeleton may inhibit recovery
altogether (Kahn et al., 2016).
Study Area
Howe Sound is in southern British Columbia, northwest of the Metro Vancouver area. It is a glacially carved
fjord along its northern section where it is fed by the Squamish River. It widens to an island archipelago in its
southern section that is an embayment of the Strait of Georgia. The fjord and embayment are separated by
a shallow (70m) submarine sill (the “Porteau Sill”) with a substrate that is largely silt and sand overlying
terminal moraine materials. The 16 km U-shaped fjord is steep-sided, narrow, and deep, averaging 3 km in
width and reaching depths of 285 m near the southern end. The outer basin is more complex, with several
large and small islands, steep rock walls, submarine ridges, channels, pinnacles and knolls. Howe Sound is
separated from the Strait of Georgia by depths less than 110m northwest of Bowen Island. Southeast of
Bowen Island is the Queen Charlotte Channel, which provides Howe Sound with a deep (approximately 240
m) connection to the Strait of Georgia.
Between 2012 and 2016 the Marine Life Sanctuary Society explored several areas of Howe Sound to identify
glass sponge presence and coverage; these areas include Porteau Sill to Ellesmere Creek; the reefs to the
east and south of Anvil Island; Lions Bay area from Brunswick Point to Kelvin Grove; the reefs south of
Halkett Point; north, west and south Bowyer Island; Dorman Point off Bowen Island; and the Passage Island
complex. These areas were targeted for exploration based on unique underwater features that had the
potential to support sponge growth, including rock walls, submarine ridges, pinnacles and knolls. Analysis of
areas closed to bottom fishing under current DFO sponge reef closures (Porteau Sill and Passage Island area)
are not included in this report.
Data Collection and Mapping
Data to map and monitor sponge aggregations in Howe Sound was collected from December 2012 to July 2016 using a 7.6m Skyliner watercraft deployed from the Lions Bay marina.
Bathymetry Mapping
Bathymetry information was collected with an on-board Garmin 300c single-beam sonar system with a
transducer connected to the stern hull of the Skyliner. Data from this system was input into the Dr. Depth
mapping software program as per the set up and calibration methodology outlined in Wilhelm & Reams
(2012)2. A bathymetry map of the area was created from the track network of survey lines made by the
Skyliner, as described by Clayton & Dennison (in press). The track network consisted of multiple survey lines
in the target area while Dr. Depth recorded the boat’s position and depth under the boat. Depths were
automatically corrected for the tidal height at the time of the assessment to display the depths relative to
chart datum. Bathymetric maps in jpeg format were overlaid onto Google Earth to interpret georeferenced
drop camera imagery.
2 Although Wilhelm and Reams (2012) provides procedures for a Hummingbird 365i sonar system, the methodology is similar for the Garmin 300c.
Marine Life Sanctuaries Society 5
Imagery Data Collection
The drop-camera system was towed along transects at each site. The path of each transect was entirely
under the influence of winds and currents while the towing vessel remained in neutral. The drop camera
position, time, and date information was used to identify approximate location and duration of
observations. A positional error occurred when the camera was not directly below the boat; this was not
measured or corrected for due to logistical difficulties in determining the angle of the camera in relation to
the boat.
The drop camera system was equipped with a low-resolution digital video camera (unknown make and
model) that recorded continuous video with position (latitude and longitude), time, and date stamp. Starting
in May of 2015, high resolution video was also collected with a Sony 1080p camera mounted to the drop
camera system. Six 850 lumen LED lights were used to illuminate the field of view for video capture.
Access to sponge aggregations less than 30m in depth was possible using recreational SCUBA divers,
allowing for monitoring of these reefs with hand-held video and still camera. Six technical tri-mix SCUBA
dives on sponge bioherms with video documentation were also carried out in 2015 to maximum depths of
84m (Dennison, 2015). Positional information for SCUBA surveys was estimated from the entry point,
underwater heading, and depth range of the dive.
Other Sources of Information
In areas where underwater video data were unavailable for this report, the location, extent, and nature of
the sponge aggregations were provided by Glen Dennison and Jeff Marliave (Tables 1 & 2). Both Glen and
Jeff had reliable, local knowledge of the aggregations identified.
Data Analysis
Sponge aggregations were identified and described based on visual information from a combination of drop
camera transects and SCUBA surveys. Each location had varying levels of information available based on the
number and length of drop camera transects, and SCUBA accessibility.
For the purpose of this study, a sponge aggregation was defined as an area where bioherm growth was
observed, where there was moderate to dense sponge growth on rock, or where there was a historical
bioherm that no longer had live growth. The latter areas are important to highlight as they may represent
areas for potential recovery of sponge aggregations in the future.
The drop-camera, SCUBA, and bathymetric data were used to identify the location and describe the nature,
landform, and depth range of the sponge aggregations. General observations of sponge condition, species
associations, and evidence of damage were also noted.
Nature
The nature of sponge aggregations in this report refers to the sponge growth form and dominant species.
Sponge growth on rock is interpreted as “sponge garden”. Sponge growth without any evidence of rock and
where it was surrounded by silt and/or other sponge was interpreted as “sponge bioherm”. Where there
was only dead sponge in dense aggregations surrounded by and covered in silt, the aggregation was referred
to as “dead sponge bioherm”. This report focused on aggregations of A. vastus and H. calyx. While the Boot
Sponge (Rhabdocalyptus dawsoni) has also been observed to form aggregations in Howe Sound (Glen
Dennison & Jeff Marliave, personal communications) and may also warrant protection, they were beyond
the scope of this analysis.
Marine Life Sanctuaries Society 6
Depth range
Depth range was interpreted from bathymetry maps overlaid with drop camera observations and reported
as meters below chart datum.
Landform
Landform refers to the physical character of the submarine feature on which sponge aggregations were
observed. Pinnacles were used to describe projections that elevate to a pointed, rounded, or flat topped
submersed peak and that drop gradually or precipitously around all sides to depths usually > 90m. These
were often connected with a saddle or isthmus to each other and/or to the shoulder of a slope that
descended from mainland or island shorelines. Walls were described as steep, rocky slopes (generally > 60°)
that may be sheer vertical or sloping rock wall or with several benches. Slopes from 20° to 60° were
described as moderate, and <20° were described as gentle.
Aggregation Description and Observations
Sponge coverage was described based on a qualitative interpretation of imagery from the drop camera
transects where available. This involved recording the start and end points of sponge observations from
each transect onto bathymetric maps overlaid on Google Earth. Also noted were areas of sponge coverage
based on classifications of none (bare substrate), sparse (<5%), moderate (5-50%), and dense (> 50%)
sponge coverage along drop camera transects. This was done over course scales that ranged from 10-20m
along the transect, depending on how homogenous the density classification was in each section. The
results of this assessment were used to describe each aggregation and contributed to the delineation of
aggregation boundaries.
Also noted with the drop camera footage and from imagery taken by SCUBA divers were observations of
interest. This included species/habitat associations, sponge condition, and evidence of damage. Images used
in this report include video frames from drop camera or other video sources and still photos. Contributors of
these images included Glen Dennison, Diane Reid, Adam Taylor, Chris Straub, and the Vancouver Aquarium.
Boundaries
Each sponge aggregation was delineated with a four-sided polygon that hugged the known or estimated
perimeter of each sponge aggregation. Known sponge locations were determined directly from the
georeferenced drop camera video footage. Boundary delineation was accomplished by first drawing the
perimeter of the sponge aggregation using the “path” feature in Google Earth that encompassed all sponge
points (sparse to dense). The four-sided polygon was then drawn to encompass this inner, more complex
polygon. The corners of the four-sided polygon are presented in decimal-degrees latitude and longitude
interpreted from Google Earth. The corners of these polygons are provided in this report and presented in a
large-scale view of the aggregation over a CHS Chart 3526 (Howe Sound) base map and individually on
bathymetric maps created using Dr. Depth, where available.
At times, drop camera information was unavailable or did not encompass the entire landform where an
aggregation was detected. Where drop camera information was unavailable but an aggregation was known
to occur (i.e. for aggregations D1-c, LB-a, and SB-c), the boundaries were delineated based on the depth
range of nearby aggregations and encompassed the entire landform on which they occurred. Where drop
camera information did not encompass the entire landform on which an aggregation was observed, the
boundaries were extrapolated to include the remaining landform at the range of depths where aggregations
had been observed on the same landform. Determining whether the landform was the same or a new
landform was a subjective process that assessed patterns in the bathymetry. For example, a discrete
Marine Life Sanctuaries Society 7
landform could include a series of knolls along the same ridge, a pinnacle separated from another or an
island shoulder by an isthmus, and an island shoulder or ridge along a similar aspect and depth range.
Figure 3 provides an example of the geo-referenced drop camera footage used to interpret benthic features
and sponge aggregation boundaries on a bathymetric map. The push pins indicate observation points
collected from drop camera footage, and are colour-coded based on sponge presence, absence, and density.
Results
Out of the 19 known sponge aggregations in Howe Sound, 13 were assessed from six general areas. The
remaining six known aggregations are currently protected within Strait of Georgia and Howe Sound Sponge
Reef Conservation Areas that prohibit bottom contact fishing activities and are not included in this report.
An overview map of sponge aggregations presented in this report in relation to current levels of protection
is provided in Figure 4.
Nature and Landform
All living sponge aggregations were dominated by the reef building sponge, A. vastus, many with less
frequent observations of the other reef building sponge, H. calyx. All sponge observations occurred where
there was moderate to high relief and/or atop benches and pinnacles exposed to currents. The overall depth
range of these aggregations was from 20 to 122 m, with the shallowest being the sponge garden at East
Defence Island (DI-b), and the deepest at the South Bowyer complex (SB-a) and Kelvin Grove Bioherm (LB-c)
(Table 1).
Aggregations included sponge bioherm, sponge garden, dead sponge bioherm, and a complex of both
bioherm and garden. All living aggregations exhibited patchy growth, encompassing both dense areas of
sponge growth and large expanses of non-living substrate, interspersed with patches of moderate to dense
growth. Many areas had sponges that were well established with many oscula. One area, the Ellesmere
Creek bioherm, had expanses of dead sponge reminiscent of a once-thriving bioherm (Glen Dennison,
personal communication). Substrate types included mud, silt, bedrock, and dead sponge. Also observed
Figure 3: Examples of tools used in aggregation description and boundary delineation. To the left is a screen shot of a drop camera frame from the South Bowyer (SB-a) aggregation with position, time, and date stamp and transect code. T right is a bathymetric map created using Dr. Depth of the SB-a aggregation. Transect endpoints are indicated with a circled “T” and connected with black lines. Colour-coded sponge observations are indicated with push-pins (blue = dense sponge; light blue = low moderate to sparse sponge; green = no sponge) interpreted from drop camera footage. Sponge aggregation boundary is delineated in red.
Marine Life Sanctuaries Society 8
were sponges that appeared to be damaged from human activities, and included broken fragments of
sponge and sheared sponge. Figure 5 provides examples of some of these observations.
Figure 4: Howe Sound glass sponge aggregations. Adapted from CHS Chart 3526.
Marine Life Sanctuaries Society 9
Species Observations
Species commonly found associated with the sponge aggregations included rockfish (Sebastes spp.) which were
often taking refuge or foraging in the folds and oscula of the living sponge. Quillback rockfish (S. magister) was
the most common rockfish species encountered; other species included Yelloweye (S. ruberrimus), Redstriped
(S. prorigor), Greenstriped (S. elongatus), and Puget Sound (S. emphaeus) rockfish. The most abundant animals
observed were the squat lobster (Munida quadrispina) and Sponge eulids (Eualus spp.). Other species of
commercial, recreational, and aboriginal fisheries interest observed in and near the sponge aggregations
included Dungeness (Metacarcinus magister) and Grooved tanner crabs (Chionoecetes tanneri), prawns
(Pandalus spp), the Giant Red sea cucumber (Parastichopus californicus), and Pacific lingcod (Ophiodon
elongatus).
Indicators of ecological relationships between sponges and other species were also observed. This included
gravid and juvenile rockfish, Pacific lingcod nest-guarding behaviour by males, foraging by rockfish and Red rock
crabs, mating Tanner crabs, and presence of the predatory nudibranch, the Freckled sea lemon (Anisodoris
lentiginosa). Observations of sponge community associations are provided in Figure 6.
Figure 5: Examples of broken (A) and sheared (B) sponge observations at Halkett Pinnacle (HP) and East Defence (DI-b) aggregations, respectively. Photo credits: (A) by Adam Taylor, (B) by the Vancouver Aquarium.
Marine Life Sanctuaries Society 10
Figure 6: Examples of sponge community associations documented by divers at some of the aggregations described in this report. A: young-of-the-year Yelloweye rockfish and two Squat lobsters at the Halkett bioherm (HP). B: Juvenile Yelloweye rockfish at Dorman Point aggregation (DP). C: Mating Tanner crabs at Halkett bioherm. D: Pacific lingcod guarding eggs at the Halkett sponge garden. E: Gravid Quillback rockfish at the Halkett bioherm. Note fishing line in this photo. F: Freckled sea lemon, a predator of glass sponge on the Halkett aggregation. Photo Credits: Photos A & F: Adam Taylor. Photos B, C, D, & E: Diane Reid.
Marine Life Sanctuaries Society 11
Assessment Summary
Table 1 provides a summary of the nature, depth range, and landform of the 13 aggregations described in this
report. Six of the aggregations were bioherms, one was sponge garden, five were sponge complexes with both
sponge garden and bioherm, and one was entirely made up of the dead biogenic sponge.
Table 2 summarizes the assessment activities that provided information for this report. Seven aggregations had
drop camera information available, and another four also were assessed with drop camera, though the footage
was not available for this report. Two aggregations were not assessed with drop camera; instead confirmation of
the aggregations was provided by individuals who had accessed them using SCUBA. Of the 13 aggregations, 12
had bathymetry information available for this report. The locations and nature of all aggregations without drop-
camera footage available were based on information provided by Glen Dennison (aggregations DI-a, AI-b, LB-a,
and SB-b) and Jeff Marliave (Aggregation DI-b).
Table 3 provides a summary of drop camera information and sampling effort where drop camera information
was available for this report.
The following sections provide a summary of the observations of sponge aggregations in Howe Sound. Overview
and inset maps are provided for each general assessment area and the associated sponge aggregation
boundaries along with a short description and images.
Table 1: Summary of the nature, landform and depth range of sponge aggregations in this report.
Aggregation
Code Aggregation Name Nature1
Landform Type
Min Depth
(m)
Max Depth
(m)
DI-a Ellesmere Creek Bioherm (dead) pinnacle n/a n/a
DI-b East Defence Island Complex submerged island shoulder / rock wall 20 40
DI-c
East Defence Island
Pinnacle Bioherm pinnacle 33 n/a
AI-a Anvil East Bioherm Bioherm submerged reef with single knoll 80 105
AI-b North Christie Garden
small knoll off of submerged island
shoulder 32 n/a
AI-c Christie / Pam Rocks Complex
multiple knolls atop a submerged
inter-island ridge / rock wall 30 80
LB-a Brunswick Bioherm Bioherm pinnacle n/a n/a
LB-b Lions Bay Bioherm Bioherm pinnacle 73 95
LB-c Kelvin Grove Bioherm Bioherm submerged island with multiple knolls 76 121
HP Halkett West Pinnacle Complex pinnacle 30 90
SB-a South Bowyer Complex submerged island shoulder / rock wall 80 122
SB-b
Southern-south
Bowyer Bioherm submerged island shoulder / rock wall n/a n/a
DP Dorman Point Complex pinnacle / rock wall 42 601 All aggregations with living sponge were dominated by A. vastus
Marine Life Sanctuaries Society 12
Table 2: Sponge aggregation discoveries and assessment types. Discovery dates provided by Clayton & Dennison (in press) and through personal communications with Glen Dennison and Jeff Marliave.
Aggregation
Code Aggregation Name Discovery Date
Discovered
by1
Assessment activities
DI-a Ellesmere Creek Unknown JM, GD drop camera2
DI-b East Defence Island 2004 JM
bathymetry mapping
recreational SCUBA
Vancouver Aquarium monitoring site
DI-c East Defence Island Pinnacle November 14, 2010 GD
drop camera2
bathymetry mapping
recreational SCUBA
AI-a Anvil East Bioherm October 4, 2010 GD, LC
drop camera
bathymetry mapping
technical SCUBA
AI-b North Christie Unknown GD
bathymetry mapping
SCUBA
AI-c Christie / Pam Rocks January 31, 2010 GD
drop camera
bathymetry mapping
recreational and technical SCUBA
LB-a Brunswick Bioherm June, 2013 GD, NF
drop camera2
bathymetry mapping
LB-b Lions Bay Bioherm January, 2013 GD
drop camera
bathymetry mapping
LB-c Kelvin Grove Bioherm January, 2013 GD
drop camera
bathymetry mapping
technical SCUBA
HP Halkett West Pinnacle June 30, 1996 GD
drop camera
bathymetry mapping
recreational and technical SCUBA
Vancouver Aquarium monitoring site
DFO/MLSS larval sponge project site
SB-a South Bowyer February 22, 2014 GD
drop camera
bathymetry mapping
SB-b Southern-south Bowyer March 22, 2014 GD
drop camera2
bathymetry mapping
DP Dorman Point October 6, 2012 GD
drop camera
bathymetry mapping
recreational SCUBA1 Discoverer(s): GD = Glen Dennison, LC = Lena Clayton, NF= Nisha Forester, JM = Jeff Marliave
2 drop camera transects conducted for this aggregation, but not available at time of this report
Marine Life Sanctuaries Society 13
Table 3: A summary of sampling effort for aggregations assessed in this document.
Assessment date
/ # transects
Total
transects
Total Length
(m)
Mean
Length
(m) SD
Min
Length
(m)
Max
Length
(m)
DI-a Ellesmere Creek Bioherm unknown 2
unknown 2
- - - - -
DI-b East Defence Island n/a 0 - - - - -
DI-c
East Defence Island
Pinnacle unknown 2
unknown 2
- - - - -
AI-a Anvil East Bioherm
Sept 27, 2014 / 1
Oct 4, 2014 / 5
Oct 18, 2014 / 6
July 11, 2015 / 2
April 23, 2016 / 3 17 1706 100 85 11 348
AI-b North Christie n/a 0 - - - - -
AI-c Christie / Pam Rocks
March 21, 2015 / 2
Aug 18, 2015 / 3 5 1279 256 196 7 539
LB-a Brunswick Bioherm unknown 2
unknown 2
- - - - -
LB-b Lions Bay Bioherm
Nov 09, 2013 / 2
April 26, 2014 / 1
July 16, 2016 / 2
July 25, 2015 / 4
Oct 17, 2015 / 2
May 7, 2016 / 2 13 1720 132 129 12 503
LB-c Kelvin Grove Bioherm
April 26, 2014 / 1
May 31, 2014 / 1
June 6, 2015 / 3
June 8, 2015 / 1
June 13, 2015 / 1
Oct 10, 2015 / 4
Nov 21, 2015 / 1
Dec 12, 2015 / 1
Jan 24, 2016 / 1
Feb 6, 2016 / 1
Feb 13, 2016 / 3
Feb 27, 2016 / 3
May 7, 2016 / 1
June 25, 2016 / 2
July 16, 2016 / 2 26 5018 186 113 21 429
HP Halkett West Pinnacle
April 26, 2014*2
Dec 13, 2012*2
Sept 19, 2015*3 7 648 93 72 24 198
SB-a South Bowyer
June 27, 2015 / 3
Sept 19, 2015 / 6
Nov 21, 2015 / 3 12 956 80 49 34 196
SB-b Southern-south Bowyer unknown 2 unknown 2 - - - - -
DP Dorman Point
Nov 30, 2013 / 6
April 25, 2015 / 4 10 418 42 25 14 94
2 drop camera transects conducted for this aggregation, but not available at time of this report
Aggregation
Code Name
Drop-camera transect data1
1 All length data reflects horizontal distance from transect endpoints. Actual transect length will increase with slope and non-linear path
between endpoints.
Marine Life Sanctuaries Society 14
Area 1: Defence Islands / Ellesmere Creek
Aggregations: Ellesmere Creek (DI-a), East Defence Island (DI-b) East Defence Island Pinnacle (DI-c)
Three aggregations were identified north of the Porteau sill. No living sponge was observed in the northernmost
Ellesmere Creek aggregation (DI-a) (Glen Denison, personal communication). Two living sponge aggregations
east of the East Defence Island were identified. The aggregation on the shoulder of East Defence Island (DI-b)
was well documented by the Vancouver Aquarium (Marliave et al, 2009). The other East Defence Island Pinnacle
aggregation (DI-c) was discovered by Glen Dennison with a drop camera, and then later explored by SCUBA,
though no imagery was available for analysis in this report (Glen Dennison, personal communication).
Figure 7: Defence Islands and Ellesmere Creek sponge complexes overview and inset maps. Adapted from CHS Chart 3526.
Table 4: Coordinates for the corners of boundaries delineating Ellesmere Creek and East Defence Island sponge aggregations.
Latitude Longitude Latitude Longitude Latitude Longitude
49°35.572'N 123°15.635'W 49°34.639'N 123°16.281'W 49°34.659'N 123°16.282'W
49°35.293'N 123°15.763'W 49°34.653'N 123°16.213'W 49°34.659'N 123°16.234'W
49°35.281'N 123°15.283'W 49°34.728'N 123°16.243'W 49°34.711'N 123°16.248'W
49°35.472'N 123°15.243'W 49°34.713'N 123°16.313'W 49°34.706'N 123°16.292'W
D1-c: East Defence Island pinnacle D1-a: Ellesmere Creek bioherm (dead) D1-b: East Defence Island
Marine Life Sanctuaries Society 15
Ellesmere Creek bioherm: Aggregation DI-a
Nature: Dead sponge bioherm, silted over (Glen
Dennison, personal communication).
Depth Range: n/a
Landform: On a pinnacle along the west margin
of Howe Sound fjord 2km north of the Porteau
sill. Pinnacle width is approximately 380m over
an undulating, gently sloped peak at
approximately 27m depth. The pinnacle
descends to greater than 100 m along all sides
except the northwest, where it is connected to
the mainland foreshore along an isthmus of 73m
depth.
Description:
• This northernmost aggregation was made up of broken and settled dead sponge skeleton (Glen Dennison,
personal communication).
• Leys et al. (2004) reported large accumulations of dead sponge skeleton from transcripts and photos from a
1982 submersible dive in this area3.
Observations:
n/a
Assessment Type:
Initially discovered with drop camera by Glen Dennison (personal communication). No video was available for
this report.
3 Transcripts from the 1982 PISCES submersible dives in “Inner Basin” area, which coincided with the general area of the Ellesmere Creek bioherm (DI-a), indicated large numbers of dead sponge (Leys et al. 2004).
Figure 8: Ellesmere Creek bioherm (DI-a) map adapted from CHS Chart 3526.
Marine Life Sanctuaries Society 16
East Defence Island: Aggregation DI-b
Nature: Living sponge garden and bioherm
complex, dominated by A. vastus.
Depth Range: 20-40m
Landform: This sponge complex occurred on
a ridge and adjacent rock wall. The ridge
runs northwest from the northeastern
shoulder of East Defence Island, and the rock
wall descends west of this ridge. This area is
separated from a small pinnacle to the
northeast where aggregation D1-c is located
by a shallow isthmus at 60m depth.
Description:
• The sponge bioherm followed along the
ridge. The sponge garden grew to the
west of the ridge along a rock wall.
• Actively monitored by the Vancouver
Aquarium for sponge growth and water
quality, including temperature4.
Observations:
• The sponge bioherm of aggregation DI-b was made up entirely by A. vastus during a biodiversity study by
Marliave et al. (2009). H. calyx was not found.
• The same study reported 15 species observations on this aggregation, including Yelloweye and Quillback
rockfish, Pacific lingcod, and a nudibranch that predates on glass sponges – the Freckled sea lemon,
Anisodoris lentiginosa.
• Coverage of sponge in the bioherm was dense; patches
with bare mud were infrequent and relatively small.
• Broken and sheared off sponges were observed at this
location.
Assessment Type: Information on description and
observations obtained from a study by the Vancouver
Aquarium (Marliave et al., 2009), photos and video provided
by Vancouver Aquarium staff and on a Vancouver Aquarium
web page4 (Vancouver Aquarium, 2017), and personal
communications from Vancouver Aquarium staff.
4 https://www.vanaqua.org/act/research/howe-sound-group/sponges
Figure 9: East Defence Island aggregation (DI-b). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Figure 10: Damage to A. vastus at DI-b; clean cut indicates it was likely caused by weighted fishing line. Photo taken Nov. 28, 2016: by April 2017, the sheared fragment had fused with the parent sponge (Jeff Marliave, personal communication). Photo provided courtesy of the Vancouver Aquarium.
Marine Life Sanctuaries Society 17
East Defence Island Pinnacle: Aggregation DI-c
Nature: Living sponge bioherm.
Depth Range: 33m at shallowest point.
Landform: This aggregation occurred east of
DI-b, on a small (40m wide) pinnacle that
ranges from 30m-90m depth. The pinnacle is
separated from the shoulder of East Defence
Island by a narrow isthmus of 60m depth.
Description:
n/a
Observations:
n/a
Assessment Type: This aggregation was
initially discovered with drop camera and
dove with SCUBA by Glen Dennison
(personal communication). No video was
available for this report, and maximum
depth is not determined. The DI-c
aggregation boundary was delineated to
include the top of the pinnacle to a depth of approximately 40m (based on maximum depth of nearby DI-b).
Figure 11: East Defence Island pinnacle aggregation (DI-c). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Marine Life Sanctuaries Society 18
Area 2: Anvil Island
Aggregations: Anvil East Bioherm (AI-a), North Christie (AI-b), Christie/Pam Rocks (AI-c)
Several sponge aggregations have been discovered around Anvil Island. The Anvil East Bioherm (AI-a), off the
northeastern shoreline along a submerged reef, was the most significant aggregation with an estimated size of 6
ha. The remaining known aggregations were found along the Christie Islet-Pam Rock island chain that trails
south from the southern tip of Anvil Island. North of Christie Islet is a shallow reef that supports a small sponge
garden (North Christie, AI-b). The area between Christie Islet and Pam Rocks has five small knolls, at least two of
which are known to support glass sponge aggregations (Christie/Pam Rocks, AI-c); the others are yet to be
surveyed. Areas east and southeast of Pam Rocks were also surveyed with drop camera, and found to lack any
sponge reef coverage, living or dead.
Figure 12: Anvil Island sponge complexes overview and inset maps. Adapted from CHS Chart 3526.
Latitude Longitude Latitude Longitude Latitude Longitude
49°32.756'N 123°17.356'W 49°30.157'N 123°18.121'W 49°29.812'N 123°18.043'W
49°32.607'N 123°17.353'W 49°30.140'N 123°18.116'W 49°29.559'N 123°17.970'W
49°32.581'N 123°17.009'W 49°30.142'N 123°18.086'W 49°29.660'N 123°17.701'W
49°32.749'N 123°16.992'W 49°30.158'N 123°18.092'W 49°29.958N 123°17.961'W
A1-c: Christie / Pam RocksAI-a Anvil East bioherm AI-b: North Christie
Table 5: Coordinates for the corners of boundaries delineating Anvil Island sponge aggregations.
Marine Life Sanctuaries Society 19
Anvil East Bioherm: Aggregation AI-a
Nature: Living sponge bioherm,
dominated by A. vastus.
Depth Range: 80-105m
Landform: This aggregation occurred
along a deep reef with a north/south
exposure and a small pinnacle on the
western side. The eastern edge of the
reef drops steeply to over 150m in
depth. An isthmus of 90m connects the
reef to eastern shoulder of Anvil Island.
Description:
• Dense sponge growth was
concentrated on and around the
pinnacle and the northerly reef
margin.
• The densest growth occurred
between 80-100m depth.
• Patches of dense, sparse, and no growth occurred throughout the reef.
Observations:
• The sponge appeared healthy, with extensive columns and large oscula, dense and continuous growth in
some areas, and patchy growth in other areas.
• A large school of rockfish (Sebastes spp., over 40 individuals) was observed in the water column over the
pinnacle during an October 18, 2014 drop camera transect.
• Broken sponge was observed at this location.
• A derelict prawn trap on sponge bioherm was also observed.
Assessment Type: Assessed with drop camera and technical SCUBA.
Figure 13: Anvil East aggregation (AI-a). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Figure 14: Clayton bioherm with dominant Aphrocallistes vastus in foreground and the rarer Heterchone calyx, top right.
Figure 15: Screen capture from Clayton bioherm showing dense live sponge on dead sponge. The Squat lobster is abundant here.
Marine Life Sanctuaries Society 20
North Christie: Aggregation AI-b
Nature: Living sponge garden growing on
rock, dominated by A. vastus.
Depth Range: 32m at shallowest point.
Landform: This small sponge garden was
perched on a shallow (32m) rock knoll with
an eastern exposure next to a steep wall.
The knoll is located at the northeastern
terminal of a submerged rocky ridge that
extends north from Christie Islet.
Description:
• Located within the Pam Rock Rockfish
Conservation Area (RCA), Pacific
Fisheries Management Area (PFMA) 28-
4.
Observations:
• There were no additional sponge
aggregations found by Glen Dennison
immediately west of the AI-b aggregation
along the ridge crest.
Assessment Type: Initially discovered with drop camera by Glen Dennison (personal communication). No video
was available for this report.
Figure 16: North Christie aggregation (AI-b). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Marine Life Sanctuaries Society 21
Christie/Pam Rocks: Aggregation AI-c
Nature: Living sponge bioherm on pinnacles
and sponge garden on wall, dominated by A.
vastus, also infrequent but established
growth of H. calyx.
Depth Range: 30-80m
Landform: AI-c was located along multiple
knolls that projected from a submerged
ridge between Christie Islet and Pam Rocks
and along the southeast shoulder of Christie
Islet. The ridge drops steeply to the east, and
is gently sloping to the west.
Description:
• Located within the Pam Rock RCA, PFMA
28-4.
• Sponge grew in dense patches along the ridge, interspersed with areas of dead sponge.
• Occasional dense growths of sponge on
rock wall occurred to the east.
• Sponge garden coverage was moderate
to dense along the submerged shoulder
of southeast Christie Islet.
• A temperature logger was installed at AI-c in 2014 and is actively maintained by MLSS.
Observations:
• This bioherm had healthy, dense and continuous growth of A. vastus and H. calyx.
• Rockfish species observed on this reef included Quillback, juvenile Yelloweye, and Redstriped.
• Sheared sponge along the pinnacle indicated damage from weighted fishing line.
Assessment Type: This complex was documented with drop camera and recreational and technical SCUBA.
Monitoring was concentrated along the eastern wall and adjacent peaks along the reef. Although the western
margin of the reef has not yet been thoroughly surveyed, it was included in the boundary based on the potential
for sponge. Potential was based on proximity of the nearby known sponge aggregation and depth range.
Figure 17: Christie / Pam Rocks aggregation (AI-c). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Figure 19: A Dungeness crab (Metacarcinus magister) perched on A. vastus.
Figure 18: A healthy example of Heterochone calyx.
Marine Life Sanctuaries Society 22
Area 3: Lions Bay
Aggregations: Brunswick Bioherm (LB-a), Lions Bay Bioherm (LB-b), Kelvin Grove Bioherm (LB-c)
There were three sponge bioherm discoveries in the Lions Bay area. The Brunswick Bioherm (LB-a) was located
approximately 350m northwest of Brunswick Bay, north of Lions Bay. The Kelvin Grove Bioherm (LB-c) was found
adjacent to the coastline between Lions Bay and Kelvin Grove. The Lions Bay Bioherm (LB-b) was approximately
1.2 km offshore on a submerged island connected to the Kelvin Grove pinnacle along a deep isthmus.
Table 6: Coordinates for the corners of boundaries delineating Lions Bay sponge aggregations.
Latitude Longitude Latitude Longitude Latitude Longitude
49°28.324'N 123°15.062'W 49°27.126'N 123°15.512'W 49°26.973'N 123°15.028'W
49°28.397'N 123°14.886'W 49°27.145'N 123°15.243'W 49°26.981'N 123°14.676'W
49°28.475'N 123°14.959'W 49°27.543'N 123°15.379'W 49°27.270'N 123°14.660'W
49°28.420'N 123°15.154'W 49°27.481'N 123°15.688'W 49°27.260'N 123°15.016'W
LB-a: Brunswick bioherm LB-b: Lions Bay Seamount LB-c: Kelvin Grove Seamount
Figure 20: Lions Bay sponge complexes overview and inset maps. Adapted from CHS Chart 3526.
Marine Life Sanctuaries Society 23
Brunswick Bioherm: Aggregation LB-a
Nature: Living sponge bioherm dominated by
A. vastus.
Depth Range: n/a
Landform: This bioherm was found on a
submarine knoll that reaches 90m at its
shallowest point.
Description:
n/a
Observations:
• Glen Dennison noted this bioherm was
healthy at the time of discovery (2013).
Assessment Type: Initially discovered with
drop camera by Glen Dennison (personal
Communication). No drop camera video was
available for this report. Figure 21: Brunswick Bioherm aggregation (LB-a). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Marine Life Sanctuaries Society 24
Lions Bay Bioherm: Aggregation LB-b
Nature: Living sponge bioherm, dominated
by A. vastus.
Depth range: 73-95m
Landform: Pinnacle with two projections
along a north-south alignment reaching
minimum depths of 89m and 73m,
respectively. Connected to Kelvin Grove
bioherm by a deep isthmus of 145m to the
southeast. Steep rock wall along westerly
margin to over 200m in depth. Steep to
moderate slopes along northern and eastern
margins.
Description:
• Densest growth of healthy sponge bioherm was observed along a north/south strip between 70-120m in depth.
Observations:
• A very patchy aggregation, with many dense areas where no seabed visible and many expanses of mud or
silted over, dead sponge.
Assessment Type: This bioherm was documented entirely with drop camera.
Figure 24: Areas of dead sponge of unknown demise on the Lions Bay bioherm.
Figure 23: Screen capture showing position of sponge along the Lions Bay bioherm
Figure 22: Lions Bay Bioherm aggregation (LB-b). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Marine Life Sanctuaries Society 25
Kelvin Grove Bioherm: Aggregation LB-c
Nature: Living sponge bioherm, dominated
by A. vastus.
Depth Range: 76-121m
Landform: This aggregation occurred along a
submerged island with four knolls, each
projecting to approximately 76m depth.
There is a steep drop off to the west with the
toe transitioning to a 400m long isthmus at
145m depth, connecting this bioherm to the
Lions Bay bioherm (LB-b). Another eastern
isthmus provides a submarine saddle to the
mainland.
Description:
• High density growth was observed along
and throughout a central northeast /
southwest axis, with patches of dense,
moderate dense, sparse and mud
throughout the remaining area.
• A UBC student project report involved
video analysis to estimate rockfish and prawn (Pandalus spp.) coverage and abundance in and around the
Kelvin Grove bioherm (Back, et al., 2016). They found a significant relationship between abundance of
rockfish and sponge coverage, but not prawns and sponge coverage.
• This aggregation had the greatest amount of effort into its assessment, with 26 drop camera transects.
Observations:
• Many areas of dense growth were interrupted by dead and broken sponge.
• Prawn traps were observed on Kelvin Grove bioherm during the 2016 commercial opening and the 2017
recreational opening.
Assessment Type: This bioherm was documented with drop camera.
Figure 25: Kelvin Grove Bioherm aggregation (LB-c). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Figure 27: Derelict prawn trap observed during a drop camera survey at the Kelvin Grove bioherm.
Figure 26: A healthy Heterochone calyx, or Goblet sponge, on the Kelvin Grove bioherm.
Marine Life Sanctuaries Society 26
Area 4: Halkett Point, Gambier Island
Aggregation: Halkett West Pinnacle (HP)
One of the few aggregations accessible at recreational diving depths, the Halkett Pinnacle garden/bioherm
complex received little assessment with drop camera. Although spatial data is scarce, it has been accessed with
SCUBA extensively by the volunteer MLSS crew and imagery from these dives provide ongoing monitoring
information. Nearby areas were assessed and were found to be devoid of sponge. These areas included a deep
pinnacle to the northeast, and a southwest to northeasterly oriented reef that spans almost 2km from the
southern tip of the pinnacle east of the Halkett West Pinnacle complex. The western pinnacle has been assessed
with drop camera, and no sponge aggregations were found here.
Figure 28: Halkett West Pinnacle sponge complex overview and inset map. Adapted from CHS Chart 3526.
Table 7: Coordinates for the corners of the boundary delineating Halkett West Pinnacle sponge complex.
Latitude Longitude
49°26.747'N 123°18.825'W
49°26.740'N 123°18.658'W
49°26.887'N 123°18.598'W
49°26.884'N 123°18.790'W
HP: Halkett West Pinnacle
Marine Life Sanctuaries Society 27
Halkett West Pinnacle: Aggregation HP
Nature: Living sponge bioherm and sponge
garden, dominated by A. vastus.
Depth range: 30-90m
Landform: The westernmost pinnacle of two
bedrock pinnacles southeast of Halkett
Point. The top of this pinnacle was elongated
along a northeast to southwest orientation
with the highest point at the southeast tip.
Precipitous drop-offs to the north and south;
east and west slopes led to the adjacent
pinnacle and Halkett Point shoulder,
respectively. Areas surrounding the rocky
pinnacle were sediment overlying bedrock.
Description:
• The southern peak of the pinnacle
hosted a dense sponge garden, deeper
adjacent and northern areas were
sponge bioherm.
• Dense growth of sponge bioherm occurred at 33m at the shallowest point.
• A logger installed at 37m depth in the bioherm has been recording temperatures since May of 2014 (MLSS
and Vancouver Aquarium). This bioherm is also the location of a current (2017) sponge larvae settlement
study by MLSS and DFO.
• A relatively flat-topped pinnacle lies to the east, not yet explored.
Observations:
• Gravid Quillback rockfish, Pacific lingcod on eggs, and juvenile Yelloweye rockfish were all regularly observed
• Evidence of fresh broken sponge in March of 2015.
Assessment Type: This bioherm was documented with drop camera and SCUBA.
Figure 29: Halkett West Pinnacle sponge aggregation (HP-a). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Figure 30: Mating Grooved tanner crabs (Chionoecetes tanneri) finding some privacy in a mound of A. vastus at the Halkett West Pinnacle. Photo by Diane Reid.
Figure 31: Quillback rockfish (Sebastes maliger) amongst schools of perch at the Halkett West Pinnacle. Photo by Adam Taylor
Marine Life Sanctuaries Society 28
Area 5: South Bowyer Island
Aggregations: South Bowyer (SB-a), Southern-south Bowyer (SB-b)
A 1.3km reef extends southward from the southern tip of Bowyer Island. Features along this reef include a
pinnacle and wall complex near the southern point of Bowyer that is accessible to SCUBA, but without any
significant sponge aggregations. Sponge aggregations do occur further south, however, at two locations along
this reef. The South Bowyer sponge aggregation (SB-a) was discovered approximately 690m south of the
southern tip of Bowyer, and the Southern-south Bowyer (SB-b) aggregation was discovered at the southern tip
of this reef (SB-b). Although collected, no drop camera information was available for the latter bioherm.
Figure 32: South Bowyer sponge complexes overview and inset map. Adapted from CHS Chart 3526.
Table 8: Coordinates for the corners of boundaries delineating South Bowyer sponge aggregations.
Latitude Longitude Latitude Longitude
49°24.679'N 123°16.150'W 49°24.352'N 123°16.156'W
49°24.605'N 123°16.144'W 49°24.357'N 123°16.112'W
49°24.577'N 123°16.025'W 49°24.448'N 123°16.095'W
49°24.656'N 123°16.016'W 49°24.449'N 123°16.185'W
SB-b: Southern-South BowyerSB-a: South Bowyer
Marine Life Sanctuaries Society 29
South Bowyer Aggregation SB-a
Nature: Living sponge bioherm, and sponge
garden on rock wall, dominated by A. vastus.
Depth range: 80-122m
Landform: The bioherm grew along a narrow
south-projecting reef. The reef drops
precipitously to the east and moderately to
the west along a rock wall. It tapers along a
moderate descending slope to the south.
The area was largely covered in sediment,
with thick mud evident along the low
gradient upper reaches of this reef.
Description:
• Growth along this bioherm was patchy in
character, with areas of dense, healthy
growth along the eastern and western
ridges.
• The reef top was mostly sand/mud.
• The steep rock wall to the east had moderate aggregations of sponge growth.
Observations:
• A diversity of life was observed, mostly in the bare patches along the reef top. Species observed included a
skate (Raja spp.), Spot prawns (Pandalus platyceros), Tanner crabs (Chionoecetes spp.), Greenstriped
rockfish, several unidentified rockfish, and a right-eye flounder (Pleuronectidae).
Assessment Type: This bioherm was documented with drop camera.
Figure 33: South Bowyer sponge aggregation (SB-a). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Figure 35: Dense bioherm growth along the eastern ridge of South Bowyer.
Figure 34: Spot prawn (Pandalus platyceros) along a bare patch near the South Bowyer bioherm.
Marine Life Sanctuaries Society 30
Southern-South Bowyer: Aggregation SB-b
Nature: Sponge bioherm
Depth range: n/a
Landform: Southern tip of reef that extends
southwards from South Bowyer Island and
drops off to the east and west. Thick mud
substrate on top and shoulder of reef, and
rock wall to the east and west.
Description:
• n/a
Observations:
• Glen Dennison found a significant glass
sponge bioherm at this location; the size
and extent are unknown.
Assessment Type: Initially discovered with
drop camera by Glen Dennison (personal
communication). No video was available for
this report. Boundary of aggregation determined based on known depth range from nearby South Bowyer
aggregation (SB-a).
Figure 36: Southern-south Bowyer sponge aggregation (SB-b). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Marine Life Sanctuaries Society 31
Area 6: Dorman Point, Bowen Island
Aggregation: Dorman Point Bioherm (DP)
Dorman Point is located along the east side of Bowen Island just south of the entrance to Snug Cove. The
Dorman Point aggregation was along a pinnacle south of Dorman Point and had both sponge bioherm and
garden.
Figure 37: Dorman Point sponge complex overview and inset map. Adapted from CHS Chart 3526.
Table 9: Coordinates for the corners of the boundary delineating the Dorman Point sponge aggregation.
Latitude Longitude
49°22.481'N 123°19.241'W
49°22.416'N 123°19.311'W
49°22.384'N 123°19.292'W
49°22.452'N 123°19.209'W
DP: Dorman Point bioherm
Marine Life Sanctuaries Society 32
Dorman Point Bioherm: Aggregation DP
Nature: This aggregation is composed of two
discrete, nearby aggregations: 1) living
sponge on rock wall and 2) living sponge
bioherm on pinnacle, both dominated by A.
vastus.
Depth range: 42m - 60m
Landform: This aggregation was associated
with a pinnacle and rock wall that extends
southeast from the submerged shoulder of
Dorman Point on Bowen Island.
Description:
• The sponge bioherm was on and around
the pinnacle.
• The sponge garden grew along the steep
drop-off southeast of the pinnacle.
Observations:
• This area receives strong currents that
funnel through the Queen Charlotte Channel.
• At least two areas of significant (greater than 100m2)
damage have been observed in the sponge bioherm, as
indicated by broken sponge debris scattered across the
seafloor, leaving large bare patches along the bioherm.
• Where growth was undisturbed, it grew thick and
productive.
• There was a high density of the Squat lobster, and schools
of perch and rockfish were common here.
Assessment Type: This aggregation was documented with
drop camera and SCUBA.
Figure 38: Dorman Point Bioherm aggregation (DP). Adapted from bathymetry map by Glen Dennison. Depth contours in feet.
Figure 40: Diver and photographer Diane Reid over the sponge bioherm of Dorman Point Bioherm. Photo by Adam Taylor.
Figure 39: Quillback and Redstriped rockfish on the Dorman Point Bioherm. Photo by Adam Taylor.
Marine Life Sanctuaries Society 33
Discussion
This project summarizes the status of knowledge on the locations and character of 13 sponge aggregations
in six different areas in Howe Sound. Six of the aggregations were bioherms, one was sponge garden, five
were sponge complexes with both sponge garden and bioherm, and one was entirely made up of the dead
biogenic sponge (Table 1). Aggregations were found using single-beam sonar bathymetry mapping, drop
camera, and SCUBA in areas exposed to currents such as pinnacles, knolls, submarine ridges, and reefs that
extended out from submerged island shoulders. All observed aggregations had areas of dense and patchy
sponge coverage dominated by the reef building sponge Aphrocallistes vastus.
Of the 13 aggregations assessed in this report, seven were mapped and delineated based on geo-referenced
drop camera footage and single-beam sonar bathymetry mapping (Table 2). The combination of these
methods provides a reliable, low cost technology to map and characterize sponge reef aggregations in Howe
Sound (Clayton & Dennison, in press). This can be a powerful tool for marine conservation planning when it
informs decisions on where and what to protect and appropriate conservation measures (Hewitt et al.,
2004; Rubidge et al., 2016). For example, a similar technology resulted in data on the distribution and
structure of benthic marine habitats in the Kent Island Group, Southeastern Australia, and was applied in the
delineation of Marine Protected Area zones there (Jordan, et al., 2005).
Limitations to the drop camera and single-beam sonar mapping system include an error in position of the
drop camera in relation to the on-board GPS receiver, gaps in data between survey lines for both drop
camera and bathymetry surveys, and a decrease in resolution with depth in single-beam sonar systems
(Brown et al., 2011; Hewitt et al., 2004). A difference in horizontal distance between the drop camera and
the on-board GPS occurs when surveying in currents and from drag while towing the camera system. While
not corrected for in this study, the error can be estimated from the length of the tow line and the angle
between the camera and boat and as outlined in Hewitt et al. (2004). Reducing error associated with gaps in
data requires higher resolution bathymetry mapping technologies (Hewitt et al., 2004). Side-scan sonar,
seismic profiling, and multi-beam swath bathymetry are examples of high-resolution technologies that have
been used to map sponge reefs along the British Columbia coast, including aggregations near Passage Island
in Howe Sound (Conway et al., 1991; Conway et al., 2007; Cook et al., 2008; Leys et al., 2004).
Review of video and still camera footage in this study identified where sponge grew at densities sufficient to
be considered an aggregation, and the limits of that growth for purposes of boundary delineation. This was a
coarse-scale assessment that summarized the density classification over broad sections of transect to
describe the general nature of each aggregation. While sufficient for this study, a finer-scale assessment of
underwater imagery could be applied to measure indicators of sponge health. This information could then
provide a baseline to assess the success of future conservation measures (Pelletier et al., 2008). A study by a
student group from UBC provides an example of a fine-scale approach to measure three ecological
indicators: the density of sponge, prawns, and rockfish. In their study, the percent coverage of sponge and
the number of prawns and rockfish was tallied within each video frame. The results were then applied to
create a sponge density map and analyzed to identify patterns in rockfish and prawn density relative to
sponge coverage. Other opportunities for fine-scale marine benthic monitoring includes the use of acoustic
and imaging techniques to measure both abiotic indicators, such as substrate type and distribution, and
biotic indicators, such as species abundance, biomass and richness (reviewed in Brown et al., 2011 and Diaz
et al., 2004).
In areas where underwater video data were unavailable for this report aggregation locations were provided
from credible sources. The locations and approximate extent of Ellesmere Creek (DI-a), Defence Island
Marine Life Sanctuaries Society 34
Pinnacle (DI-c), Brunswick Pinnacle (LB-a) and Southern-south Bowyer (SB-b) aggregations were provided by
Glen Dennison, the author of several reports on diving and sponge reefs in Howe Sound (Clayton &
Dennison, in press; Dennison, 2012, 2015). Glen is also the owner and operator of the drop camera and
bathymetry mapping system used in this study. He has been actively mapping the benthic habitats of Howe
Sound since 2011 and diving them for over 30 years. A detailed explanation of the location, extent and
nature of the East Defence Island aggregation (DI-b) was provided by Dr. Jeff Marliave, of the Vancouver
Aquarium. Dr. Marliave has been involved in numerous research SCUBA dives at this and many other
locations in Howe Sound (Marliave, 2015; Marliave et al., 2009).
The still and video imagery collected by SCUBA divers documented the status of and species associated with
the sponge aggregations. The use of video and still imagery taken by SCUBA divers is an important tool in
documenting underwater life (Mallet & Pelletier, 2014). Refinement of this type of data collection in the
future can involve the compilation of a species database that includes the general location, date, time,
depth, and photographer. Over time the information gathered can contribute to a greater understanding of
biodiversity and seasonality on the sponge aggregations, important metrics for monitoring health (Pelletier
et al., 2008).
Past studies in Howe Sound have also contributed to our knowledge of sponge aggregations. The Defence
Island East bioherm/sponge garden complex was the focus of a study of biodiversity and rockfish
recruitment by Marliave et al. (2009), who found sponge bioherms to have lower community biodiversity
and juvenile rockfish than sponge gardens. A review of the photographs and transcripts from a series of
submersible dives in the 1980s in Howe Sound by Leys et al. (2004) found that Howe Sound had the
shallowest record of A. vastus at 18m when compared to numerous inlets along the BC coast. This study also
identified large accumulations of dead sponge in the general area of the Ellesmere dead bioherm, referred
to as the “Inner Basin” on a large-scale map in their report. The potential cause of death in these sponges
was hypothesized to be from a sustained anoxia event in the fjord recorded in 1977 and 1978 (Levings &
McDaniel, 1980). Historic pollution of Howe Sound from the Woodfibre pulp mill, the Britannia Copper mine,
and excess sediment transport from mining in the Squamish watershed were also cited as factors that may
affect sponge survival in Howe Sound.
Many species observed in this study were significant to aboriginal, commercial, and recreational fisheries.
Several species of rockfish were encountered amongst the sponge aggregations that included young-of-the-
year, juvenile, sub-adult, adult, and mature (gravid) life stages. Cook et al. (2008) also found rockfish
associated with sponge reefs of the Strait of Georgia and found important linkages between sponge health
and rockfish density. Other observations of interest included Pacific lingcod caring for egg masses amongst
the sponge during the winter months, demersal crustaceans that are known prey items for rockfish (Murie,
1995; Yang & Nelson, 2000), and larger crustaceans that are the target for some fisheries, such as Red rock
crab, Tanner crabs, and prawns. These observations provide insight on the importance of sponge reefs in
sustaining species of significance to fisheries.
The occurrence of species of fisheries significance on the sponge aggregations of Howe Sound presents both
a problem and a potential solution for protection. A problem is presented when sponge reefs are targeted
by fisheries for the species they support, and in the process become damaged by fishing gear. Fishing
methods that have the potential to damage sponges are those that come into contact with the sea bed and
include hook and line, trap, and trawl. The impacts are evident, and include sheared and broken sponge
(Figure 5B & see Austin et al., 2007). These observations, along with indications that loss of sponge habitat
will also impact fisheries (Cook et al., 2008), may provide an incentive for resource users to avoid fishing in
these areas. Currently, some of these areas are targeted by fishers, though it is unknown if the users are
Marine Life Sanctuaries Society 35
aware of the damage being caused to both the sponge and the fisheries resources they support (Figure 41).
Sharing information on the location and importance of sponge aggregations in Howe Sound may reduce
fishing pressure and allow for recovery of already damaged sponge aggregations. Studies of sponge growth
after damage indicate recovery is possible (Marliave, 2015), however the recovery process can be slow and
may not be possible if damage is significant and ongoing (Austin et al., 2007).
The information presented in this report was the result of citizen science data collection by members of the
MLSS and the Vancouver Aquarium. This work is ongoing, and made possible by funders that contributed
towards the many expenses associated with mapping the benthos of Howe Sound. The value of the
information is evident from the application of these results towards a goal to protect sensitive sponge reef
habitats. Although this is the accumulation of five years of work documenting sponge reef habitats, there
are many areas in Howe Sound that have not yet been explored. These areas include the less accessible
western fringes of Howe Sound. With support from volunteers and the community, continued monitoring
and documentation of the known reefs, and inclusion of mapping of the farther unexplored reaches, MLSS
will continue to contribute to and enhance our understanding of sponge reef aggregations and distribution
in Howe Sound.
Figure 41: Photo taken in May, 2017 of several buoys marking prawn traps set over the Lions Bay aggregation, LB-b.
Marine Life Sanctuaries Society 36
Acknowledgements
Many people participated and contributed towards the knowledge, images, and data that provided the
information necessary to outline our current knowledge of sponge reefs in Howe Sound. The countless
hours of volunteer mapping and diving by Glen Dennison and his assistants, Lena Clayton and Nisha
Forester, resulted in the underwater video, bathymetry, and local knowledge used in the analysis for this
report. Sheila Byers of the MLSS, Chantelle Caron of Fisheries and Oceans, and Glen Dennison reviewed and
provided helpful feedback on earlier revisions of this report. Photographers Diane Reid, and Adam Taylor
provided images from some of the air-diveable sponge reefs, and expertly captured their beauty and
fragility. Chris Straub and Hamish Tweed and their support teams, including Top Line Charters, provided first
hand documentation of some of the deep reefs using technical SCUBA. Scientists Jeff Marliave, Jessica Shultz
and Donna Gibbs of the Vancouver Aquarium shared their knowledge, images, reports, and maps on the
Defence Islands sponge reefs. Fisheries and Oceans, care of Aleria Ladwig and Chantelle Caron, provided the
funding for this report which helped make possible this consolidation of many years of information gathered
on the sponge reefs of Howe Sound.
Marine Life Sanctuaries Society 37
References
Austin, W. C. (1999). The relationship of silicatelevels to the shallow water distribution of hexactinellids in British Columbia. Memoirs of The Queensland Museum, 44, 44–44.
Austin, W. C., Conway, K. W., Barrie, J. V., & Krautter, M. (2007). Growth and morphology of a reef-forming glass sponge, Aphrocallistes vastus (Hexactinellida), and implications for recovery from widespread trawl damage. In M. R. Custódio, G. L.-H. Muricy, E. Hajdu, & G. Muricy (Eds.), Porifera Research, Biodiversity, Innovation and Sustainability (SÉRIE LIVR, pp. 139–145). Rio De Janeiro: Museu Nacional.
Back, T., Cawley, K., Allen, K., & Lynch, S. (2016). Glass Sponge Reefs in Howe Sound, British Columbia: Assessing the relationship between live sponge coverage and populations of prawns (Pandalus spp.) and rockfishes (Sebastes spp.). Vancouver, BC.
Brown, C. J., Smith, S. J., Lawton, P., & Anderson, J. T. (2011). Benthic habitat mapping: A review of progress towards improved understanding of the spatial ecology of the seafloor using acoustic techniques. Estuarine, Coastal and Shelf Science, 92(3), 502–520.
Chu, J. W. F., Maldonado, M., Yahel, G., & Leys, S. P. (2011). Glass sponge reefs as a silicon sink. Marine Ecology Progress Series, 441, 1–14.
Clayton, L., & Dennison, G. (in press). Inexpensive Video Drift Camera Method for Exploring Seabed in Howe Sound, BC to Document Glass Sponge Reefs. The Canadian Field-Naturalist.
Conway, K. W. (1999). Fisheries and Oceans Canada Pêches et Océans Canada. Canadian Stock Assessment Secretariat Research Document 99/192. Ottawa.
Conway, K. W., Barrie, J. V., Austin, W. C., & Luternauer, J. L. (1991). Holocene sponge bioherms on the western Canadian continental shelf. Continental Shelf Research, 11(8–10), 771–790.
Conway, K. W., Barrie, J. V., Hill, P. R., Austin, W. C., & Picard, K. (2007). Mapping sensitive benthic habitats in the Strait of Georgia, coastal British Columbia: deep-water sponge and coral reefs. Geological Survey of Canada, Current Research, A-2 (January), 6.
Conway, K. W., Barrie, J. V., & Krautter, M. (2005). Geomorphology of unique reefs on the western Canadian shelf: Sponge reefs mapped by multibeam bathymetry. Geo-Marine Letters, 25(4), 205–213.
Conway, K. W., Krautter, M., Barrie, J. V., & Neuweiler, M. (2001). Hexactinellid sponge reefs on the Canadian Continental Shelf: A unique “living fossil.” Geoscience Canada, 28(2).
Cook, S. E., Conway, K. W., & Burd, B. (2008). Status of the glass sponge reefs in the Georgia Basin. Marine Environmental Research, 66, S80–S86.
Dennison, G. (2012). Diving Howe Sound Reefs and Islands.
Dennison, G. (2015). Marine Life Sanctuaries Society 2015 Howe Sound Studies Report.
Diaz, R. J., Solan, M., & Valente, R. M. (2004). A review of approaches for classifying benthic habitats and evaluating habitat quality. Journal of Environmental Management, 73(3), 165–181.
Fisheries and Oceans Canada. (n.d.). Strait of Georgia and Howe Sound Glass Sponge Reef Fishing Closures. Retrieved January 10, 2017, from http://dfo-mpo.gc.ca/oceans/ceccsr-cerceef/closures-fermetures-eng.html
Hewitt, J. E., Thrush, S. F., Legendre, P., Funnel, G. A., Ellis, J., & Morrison, M. (2004). Mapping of Marine Soft-Sediment Communities: Integrated Sampling for Ecological Interpretation. Ecological Applications, 14(4), 1203–1216.
Marine Life Sanctuaries Society 38
Jordan, A., Lawler, M., Halley, V., & Barrett, N. (2005). Seabed habitat mapping in the Kent Group of islands and its role in marine protected area planning. Aquatic Conservation: Marine and Freshwater Ecosystems, 15(1), 51–70.
Kahn, A. S., Vehring, L. J., Brown, R. R., & Leys, S. P. (2016). Dynamic change, recruitment and resilience in reef-forming glass sponges. Journal of the Marine Biological Association of the United Kingdom, 96(2), 429–436.
Kahn, A., Yahel, G., Chu, W., Tunnicliffe, V., & Leys, S. P. (2015). Benthic grazing and carbon sequestration by deep-water glass sponge reefs glass sponge reefs. Limnology and Oceanography, 0, 1–19.
Krautter, M., Conway, K. W., Barrie, J. V., & Neuweiler, M. (2001). Discovery of a “Living Dinosaur”: Globally unique modern hexactinellid sponge reefs off British Columbia, Canada. Facies, 44, 265–282.
Levings, C. D., & McDaniel, N. G. (1980). Data Report on Effects of Dissolved Oxygen Deficiency in Bottom Waters of Howe Sound: Trawl Data January 1978 to March 1979 and Oceanographic Data August 1977 to March 1979. Canada Data Report on Fisheries and Aquatic Sciences, (217).
Leys, S. P. (2013). Effects of Sediment on Glass Sponges ( Porifera , Hexactinellida ) and projected effects on Glass Sponge Reefs (Canadian Science Advisory Secretariat (CSAS) No. Research Document 2013/074). Edmonton, Ab.
Leys, S. P., & Lauzon, N. R. J. (1998). Hexactinellid sponge ecology: growth rates and seasonality in deep water sponges, 230, 111–129.
Leys, S. P., Mackie, G. O., & Reiswig, H. M. (2007). The Biology of Glass Sponges. Advances in Marine Biology (Vol. 52). Academic Press.
Leys, S. P., Wilson, K., Holeton, C., Reiswig, H. M., Austin, W. C., & Tunnicliffe, V. (2004). Patterns of glass sponge (Porifera , Hexactinellida) distribution in coastal waters of British Columbia , Canada. Marine Ecology Progress Series, 283, 133–149.
Leys, S. P., Yahel, G., Reidenbach, M. A., Tunnicliffe, V., Shavit, U., & Henry, M. (2011). The Sponge Pump : The Role of Current Induced Flow in the Design of the Sponge Body Plan. PloS One, 6(12).
Mallet, D., & Pelletier, D. (2014). Underwater video techniques for observing coastal marine biodiversity: A review of sixty years of publications (1952-2012). Fisheries Research, 154, 44–62.
Marliave, J. B. (2015). Cloud Sponge, Aphrocallistes vastus (Porifera: Hexactinellida), Fragment Healing and Reattachment. Canadian Field-Naturalist, 129(4), 399–402.
Marliave, J. B., Conway, K. W., Gibbs, D. M., Lamb, A., & Gibbs, C. (2009). Biodiversity and rockfish recruitment in sponge gardens and bioherms of southern British Columbia, Canada. Marine Biology, 156, 2247–2254.
Murie, D. J. (1995). Comparative feeding ecology of two sympatric rockfish congeners, Sebastes caurinus (Copper rockfish) and S. maliger (Quillback rockfish ). Marine Biology, 124, 341–353.
Pechenik, J. A. (1996). Biology of the Invertebrates. (M. Kemp, Ed.). Dubuque, IA: Wm. C. Brown Publishers.
Pelletier, D., Claudet, J., Ferraris, J., Benedetti-Cecchi, L., & Garcìa-Charton, J. A. (2008). Models and indicators for assessing conservation and fisheries-related effects of marine protected areas. Canadian Journal of Fisheries and Aquatic Sciences, 65(4), 765–779.
Reiswig, H. M. (2002). Family Aphrocallistidae. In J. N. A. Hooper & R. W. M. Van Soest (Eds.), Systema Porifera: A Guide to the Classification of Sponges (pp. 1282–1286). New York: Kluwer Academic/Plenum Publishers.
Marine Life Sanctuaries Society 39
Rubidge, E., Gale, K. S. P., Curtis, J. M. R., Mcclelland, E., Feyrer, L., & Robb, C. (2016). Methodology of the Pacific Marine Ecological Classification System and its Application to the Northern and Southern Shelf Bioregions, (May).
Stockner, J., Cliff, D., & Buchanan, D. B. (1977). Phytoplankton Production and Distribution in Howe Sound , British Columbia: A Coastal Marine Embayment-Fjord Under Stress. Journal of Fisheries Research Board of Canada, 34, 907–917.
Tompkins-Macdonald, G. J., & Leys, S. P. (2008). Glass sponges arrest pumping in response to sediment: Implications for the physiology of the hexactinellid conduction system. Marine Biology, 154(6), 973–984.
Vancouver Aquarium. (2017). Diving Sponge Reefs & Gardens. Retrieved February 24, 2017, from https://www.vanaqua.org/act/research/howe-sound-group/sponges
Whitney, F., Conway, K. W., Thomson, R., Barrie, V., Krautter, M., & Mungov, G. (2005). Oceanographic habitat of sponge reefs on the Western Canadian Continental Shelf. Continental Shelf Research, 25, 211–226.
Wilhelm, F., & Reams, E. (2012). Standard Operating Procedure for Bathymetric Mapping with Humminbird 365i and Dr. Depth Software (Vol. September). Moscow, Id.
Wyeth, R. C., Leys, S. P., & Mackie, G. O. (1996). Use of sandwich cultures for the study of feeding in the hexactinellid sponge Rhabdocalyptus dawsoni (Lambe, 1892). Acta Zoologica, 77(3), 227–232.
Yahel, G., Whitney, F., Reiswig, H. M., & Leys, S. P. (2007). In situ feeding and metabolism of glass sponges (Hexactinellida , Porifera ) studied in a deep temperate fjord with a remotely operated submersible, 52(1), 428–440.
Yang, M. S., & Nelson, M. W. (2000). Food Habits of the Commercially Important Groundfishes in the Gulf of Alaska in 1990, 1993, and 1996. NOAA Technical Memo. NMFS-AFSC-112.